# zhusuan.legacy¶

## Special¶

class Empirical(dtype, batch_shape=None, value_shape=None, group_ndims=0, is_continuous=None, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of Empirical distribution. Distribution for any variables, which are sampled from an empirical distribution and have no explicit density. You can not sample from the distribution or calculate probabilities and log-probabilities. See Distribution for details.

Parameters: dtype – The value type of samples from the distribution. batch_shape – A TensorShape describing the batch_shape of the distribution. value_shape – A TensorShape describing the value_shape of the distribution. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. is_continuous – A bool or None. Whether the distribution is continuous or not. If None, will consider it continuous only if dtype is a float type.
class Implicit(samples, value_shape=None, group_ndims=0, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of Implicit distribution. The distribution abstracts variables whose distribution have no explicit form. A common example of implicit variables are the generated samples from a GAN. See Distribution for details.

Parameters: samples – A Tensor. value_shape – A TensorShape describing the value_shape of the distribution. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation.

## Stochastic¶

class Normal(name, mean=0.0, _sentinel=None, std=None, logstd=None, n_samples=None, group_ndims=0, is_reparameterized=True, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of univariate Normal StochasticTensor. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. _sentinel – Used to prevent positional parameters. Internal, do not use. mean – A float Tensor. The mean of the Normal distribution. Should be broadcastable to match logstd. std – A float Tensor. The standard deviation of the Normal distribution. Should be positive and broadcastable to match mean. logstd – A float Tensor. The log standard deviation of the Normal distribution. Should be broadcastable to match mean. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. is_reparameterized – A Bool. If True, gradients on samples from this StochasticTensor are allowed to propagate into inputs, using the reparametrization trick from (Kingma, 2013). check_numerics – Bool. Whether to check numeric issues.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class FoldNormal(name, mean=0.0, _sentinel=None, std=None, logstd=None, n_samples=None, group_ndims=0, is_reparameterized=True, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of univariate FoldNormal StochasticTensor. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. mean – A float Tensor. The mean of the FoldNormal distribution. Should be broadcastable to match std or logstd. _sentinel – Used to prevent positional parameters. Internal, do not use. std – A float Tensor. The standard deviation of the FoldNormal distribution. Should be positive and broadcastable to match mean. logstd – A float Tensor. The log standard deviation of the FoldNormal distribution. Should be broadcastable to match mean. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. is_reparameterized – A Bool. If True, gradients on samples from this StochasticTensor are allowed to propagate into inputs, using the reparametrization trick from (Kingma, 2013). check_numerics – Bool. Whether to check numeric issues.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class Bernoulli(name, logits, n_samples=None, group_ndims=0, dtype=tf.int32, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of univariate Bernoulli StochasticTensor. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. logits – A float Tensor. The log-odds of probabilities of being 1. $\mathrm{logits} = \log \frac{p}{1 - p}$ n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. dtype – The value type of this StochasticTensor. Can be int (tf.int16, tf.int32, tf.int64) or float (tf.float16, tf.float32, tf.float64). Default is int32.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class Categorical(name, logits, n_samples=None, group_ndims=0, dtype=tf.int32, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of univariate Categorical StochasticTensor. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. logits – A N-D (N >= 1) float Tensor of shape (…, n_categories). Each slice [i, j,…, k, :] represents the un-normalized log probabilities for all categories. $\mathrm{logits} \propto \log p$ n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. dtype – The value type of this StochasticTensor. Can be float32, float64, int32, or int64. Default is int32.

A single sample is a (N-1)-D Tensor with tf.int32 values in range [0, n_categories).

bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class OnehotCategorical(name, logits, n_samples=None, group_ndims=0, dtype=tf.int32, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of one-hot Categorical StochasticTensor. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. logits – A N-D (N >= 1) float Tensor of shape (…, n_categories). Each slice [i, j, …, k, :] represents the un-normalized log probabilities for all categories. $\mathrm{logits} \propto \log p$ n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. dtype – The value type of this StochasticTensor. Can be int (tf.int16, tf.int32, tf.int64) or float (tf.float16, tf.float32, tf.float64). Default is int32.

A single sample is a N-D Tensor with the same shape as logits. Each slice [i, j, …, k, :] is a one-hot vector of the selected category.

bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
Discrete
OnehotDiscrete
class Uniform(name, minval=0.0, maxval=1.0, n_samples=None, group_ndims=0, is_reparameterized=True, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of univariate Uniform StochasticTensor. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. minval – A float Tensor. The lower bound on the range of the uniform distribution. Should be broadcastable to match maxval. maxval – A float Tensor. The upper bound on the range of the uniform distribution. Should be element-wise bigger than minval. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. is_reparameterized – A Bool. If True, gradients on samples from this StochasticTensor are allowed to propagate into inputs, using the reparametrization trick from (Kingma, 2013). check_numerics – Bool. Whether to check numeric issues.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class Gamma(name, alpha, beta, n_samples=None, group_ndims=0, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of univariate Gamma StochasticTensor. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. alpha – A float Tensor. The shape parameter of the Gamma distribution. Should be positive and broadcastable to match beta. beta – A float Tensor. The inverse scale parameter of the Gamma distribution. Should be positive and broadcastable to match alpha. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. check_numerics – Bool. Whether to check numeric issues.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class Beta(name, alpha, beta, n_samples=None, group_ndims=0, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of univariate Beta StochasticTensor. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. alpha – A float Tensor. One of the two shape parameters of the Beta distribution. Should be positive and broadcastable to match beta. beta – A float Tensor. One of the two shape parameters of the Beta distribution. Should be positive and broadcastable to match alpha. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. check_numerics – Bool. Whether to check numeric issues.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class Poisson(name, rate, n_samples=None, group_ndims=0, dtype=tf.int32, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of univariate Poisson StochasticTensor. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. rate – A float Tensor. The rate parameter of Poisson distribution. Must be positive. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. dtype – The value type of this StochasticTensor. Can be int (tf.int16, tf.int32, tf.int64) or float (tf.float16, tf.float32, tf.float64). Default is int32. check_numerics – Bool. Whether to check numeric issues.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class Binomial(name, logits, n_experiments, n_samples=None, group_ndims=0, dtype=tf.int32, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of univariate Binomial StochasticTensor. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. logits – A float Tensor. The log-odds of probabilities. $\mathrm{logits} = \log \frac{p}{1 - p}$ n_experiments – A 0-D int32 Tensor. The number of experiments for each sample. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. dtype – The value type of this StochasticTensor. Can be int (tf.int16, tf.int32, tf.int64) or float (tf.float16, tf.float32, tf.float64). Default is int32. check_numerics – Bool. Whether to check numeric issues.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class InverseGamma(name, alpha, beta, n_samples=None, group_ndims=0, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of univariate InverseGamma StochasticTensor. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. alpha – A float Tensor. The shape parameter of the InverseGamma distribution. Should be positive and broadcastable to match beta. beta – A float Tensor. The scale parameter of the InverseGamma distribution. Should be positive and broadcastable to match alpha. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. check_numerics – Bool. Whether to check numeric issues.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class Laplace(name, loc, scale, n_samples=None, group_ndims=0, is_reparameterized=True, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of univariate Laplace StochasticTensor. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. loc – A float Tensor. The location parameter of the Laplace distribution. Should be broadcastable to match scale. scale – A float Tensor. The scale parameter of the Laplace distribution. Should be positive and broadcastable to match loc. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. is_reparameterized – A Bool. If True, gradients on samples from this StochasticTensor are allowed to propagate into inputs, using the reparametrization trick from (Kingma, 2013). check_numerics – Bool. Whether to check numeric issues.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class MultivariateNormalCholesky(name, mean, cov_tril, n_samples=None, group_ndims=0, is_reparameterized=True, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of multivariate normal StochasticTensor, where covariance is parameterized with the lower triangular matrix $$L$$ in Cholesky decomposition $$LL^T = \Sigma$$.

See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. mean – An N-D float Tensor of shape […, n_dim]. Each slice [i, j, …, k, :] represents the mean of a single multivariate normal distribution. cov_tril – An (N+1)-D float Tensor of shape […, n_dim, n_dim]. Each slice [i, …, k, :, :] represents the lower triangular matrix in the Cholesky decomposition of the covariance of a single distribution. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. is_reparameterized – A Bool. If True, gradients on samples from this distribution are allowed to propagate into inputs, using the reparametrization trick from (Kingma, 2013). check_numerics – Bool. Whether to check numeric issues.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class MatrixVariateNormalCholesky(name, mean, u_tril, v_tril, n_samples=None, group_ndims=0, is_reparameterized=True, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of matrix variate normal StochasticTensor, where covariances $$U$$ and $$V$$ are parameterized with the lower triangular matrix in Cholesky decomposition,

$L_u \text{s.t.} L_u L_u^T = U,\; L_v \text{s.t.} L_v L_v^T = V$

See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. mean – An N-D float Tensor of shape […, n_row, n_col]. Each slice [i, j, …, k, :, :] represents the mean of a single matrix variate normal distribution. u_tril – An N-D float Tensor of shape […, n_row, n_row]. Each slice [i, j, …, k, :, :] represents the lower triangular matrix in the Cholesky decomposition of the among-row covariance of a single matrix variate normal distribution. v_tril – An N-D float Tensor of shape […, n_col, n_col]. Each slice [i, j, …, k, :, :] represents the lower triangular matrix in the Cholesky decomposition of the among-column covariance of a single matrix variate normal distribution. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. is_reparameterized – A Bool. If True, gradients on samples from this distribution are allowed to propagate into inputs, using the reparametrization trick from (Kingma, 2013). check_numerics – Bool. Whether to check numeric issues.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class Multinomial(name, logits, n_experiments, normalize_logits=True, n_samples=None, group_ndims=0, dtype=tf.int32, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of Multinomial StochasticTensor. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. logits – A N-D (N >= 1) float Tensor of shape […, n_categories]. Each slice [i, j, …, k, :] represents the log probabilities for all categories. By default (when normalize_logits=True), the probabilities could be un-normalized. $\mathrm{logits} \propto \log p$ n_experiments – A 0-D int32 Tensor or None. When it is a 0-D int32 integer, it represents the number of experiments for each sample, which should be invariant among samples. In this situation _sample function is supported. When it is None, _sample function is not supported, and when calculating probabilities the number of experiments will be inferred from given, so it could vary among samples. normalize_logits – A bool indicating whether logits should be normalized when computing probability. If you believe logits is already normalized, set it to False to speed up. Default is True. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. dtype – The value type of this StochasticTensor. Can be int (tf.int16, tf.int32, tf.int64) or float (tf.float16, tf.float32, tf.float64). Default is int32.

A single sample is a N-D Tensor with the same shape as logits. Each slice [i, j, …, k, :] is a vector of counts for all categories.

bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class UnnormalizedMultinomial(name, logits, normalize_logits=True, group_ndims=0, dtype=tf.int32, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of UnnormalizedMultinomial StochasticTensor. UnnormalizedMultinomial distribution calculates probabilities differently from Multinomial: It considers the bag-of-words given as a statistics of an ordered result sequence, and calculates the probability of the (imagined) ordered sequence. Hence it does not multiply the term

$\binom{n}{k_1, k_2, \dots} = \frac{n!}{\prod_{i} k_i!}$

See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. logits – A N-D (N >= 1) float Tensor of shape […, n_categories]. Each slice [i, j, …, k, :] represents the log probabilities for all categories. By default (when normalize_logits=True), the probabilities could be un-normalized. $\mathrm{logits} \propto \log p$ normalize_logits – A bool indicating whether logits should be normalized when computing probability. If you believe logits is already normalized, set it to False to speed up. Default is True. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. dtype – The value type of this StochasticTensor. Can be int (tf.int16, tf.int32, tf.int64) or float (tf.float16, tf.float32, tf.float64). Default is int32.

A single sample is a N-D Tensor with the same shape as logits. Each slice [i, j, …, k, :] is a vector of counts for all categories.

bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
BagofCategoricals
class Dirichlet(name, alpha, n_samples=None, group_ndims=0, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of Dirichlet StochasticTensor. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. alpha – A N-D (N >= 1) float Tensor of shape (…, n_categories). Each slice [i, j, …, k, :] represents the concentration parameter of a Dirichlet distribution. Should be positive. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. check_numerics – Bool. Whether to check numeric issues.

A single sample is a N-D Tensor with the same shape as alpha. Each slice [i, j, …, k, :] of the sample is a vector of probabilities of a Categorical distribution [x_1, x_2, … ], which lies on the simplex

$\sum_{i} x_i = 1, 0 < x_i < 1$
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class BinConcrete(name, temperature, logits, n_samples=None, group_ndims=0, is_reparameterized=True, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of univariate BinConcrete StochasticTensor from (Maddison, 2016). It is the binary case of Concrete. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. temperature – A 0-D float Tensor. The temperature of the relaxed distribution. The temperature should be positive. logits – A float Tensor. The log-odds of probabilities of being 1. $\mathrm{logits} = \log \frac{p}{1 - p}$ n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. is_reparameterized – A Bool. If True, gradients on samples from this StochasticTensor are allowed to propagate into inputs, using the reparametrization trick from (Kingma, 2013). check_numerics – Bool. Whether to check numeric issues.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
BinGumbelSoftmax
class ExpConcrete(name, temperature, logits, n_samples=None, group_ndims=0, is_reparameterized=True, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of ExpConcrete StochasticTensor from (Maddison, 2016), transformed from Concrete by taking logarithm. See StochasticTensor for details.

Parameters: temperature – A 0-D float Tensor. The temperature of the relaxed distribution. The temperature should be positive. logits – A N-D (N >= 1) float Tensor of shape (…, n_categories). Each slice [i, j, …, k, :] represents the un-normalized log probabilities for all categories. $\mathrm{logits} \propto \log p$ n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. is_reparameterized – A Bool. If True, gradients on samples from this StochasticTensor are allowed to propagate into inputs, using the reparametrization trick from (Kingma, 2013). check_numerics – Bool. Whether to check numeric issues.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
ExpGumbelSoftmax
class Concrete(name, temperature, logits, n_samples=None, group_ndims=0, is_reparameterized=True, check_numerics=False, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of Concrete (or Gumbel-Softmax) StochasticTensor from (Maddison, 2016; Jang, 2016), served as the continuous relaxation of the OnehotCategorical. See StochasticTensor for details.

Parameters: temperature – A 0-D float Tensor. The temperature of the relaxed distribution. The temperature should be positive. logits – A N-D (N >= 1) float Tensor of shape (…, n_categories). Each slice [i, j, …, k, :] represents the un-normalized log probabilities for all categories. $\mathrm{logits} \propto \log p$ n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. is_reparameterized – A Bool. If True, gradients on samples from this StochasticTensor are allowed to propagate into inputs, using the reparametrization trick from (Kingma, 2013). check_numerics – Bool. Whether to check numeric issues.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
GumbelSoftmax
class Empirical(name, dtype, batch_shape, n_samples=None, group_ndims=0, value_shape=None, is_continuous=None, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of Empirical StochasticTensor. For any inference it is always required that the variables are observed. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. dtype – The value type of samples from the distribution. batch_shape – A TensorShape describing the batch_shape of the distribution. value_shape – A TensorShape describing the value_shape of the distribution. is_continuous – A bool or None. Whether the distribution is continuous or not. If None, will consider it continuous only if dtype is a float type. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.
class Implicit(name, samples, value_shape=None, group_ndims=0, n_samples=None, **kwargs)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The class of Implicit StochasticTensor. This distribution always sample the implicit tensor provided. See StochasticTensor for details.

Parameters: name – A string. The name of the StochasticTensor. Must be unique in the BayesianNet context. samples – A N-D (N >= 1) float Tensor. value_shape – A list or tuple describing the value_shape of the distribution. The entries of the list can either be int, Dimension or None. group_ndims – A 0-D int32 Tensor representing the number of dimensions in batch_shape (counted from the end) that are grouped into a single event, so that their probabilities are calculated together. Default is 0, which means a single value is an event. See Distribution for more detailed explanation. n_samples – A 0-D int32 Tensor or None. Number of samples generated by this StochasticTensor.
bn

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
cond_log_p

The conditional log probability of the StochasticTensor, evaluated at its current value (given by tensor).

Returns: A Tensor.
dist
The distribution followed by the StochasticTensor.
Returns: A Distribution instance.
distribution

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The distribution followed by the StochasticTensor.

Returns: A Distribution instance.
dtype

The sample type of the StochasticTensor.

Returns: A DType instance.
get_shape()

Alias of shape.

Returns: A TensorShape instance.
is_observed()

Whether the StochasticTensor is observed or not.

Returns: A bool.
log_prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the log probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The log probability value.
name

The name of the StochasticTensor.

Returns: A string.
net

Warning

Deprecated in 0.4, will be removed in 0.4.1.

The BayesianNet where the StochasticTensor lives.

Returns: A BayesianNet instance.
prob(given)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Compute the probability density (mass) function of the underlying distribution at the given value.

Parameters: given – A Tensor. A Tensor. The probability value.
sample(n_samples)

Warning

Deprecated in 0.4, will be removed in 0.4.1.

Sample from the underlying distribution.

Parameters: n_samples – A 0-D int32 Tensor. The number of samples. A Tensor.
shape

Return the static shape of this StochasticTensor.

Returns: A TensorShape instance.
tensor

The value of this StochasticTensor. If it is observed, then the observation is returned, otherwise samples are returned.

Returns: A Tensor.